any graph transformation will result in the loss of batch information. for a workaround , you may should use set_batch_num_nodes and set_batch_num_edges

https://www.cnblogs.com/liyinggang/p/13360917.html

https://zhuanlan.zhihu.com/p/406833147

https://blog.csdn.net/wufeil7/article/details/107106299

1.使用内置的消息传递api

比如GraphConv

2.实现自己的消息传递策略

Write your own GNN module

https://docs.dgl.ai/tutorials/blitz/3_message_passing.html

Message Passing APIs

https://docs.dgl.ai/guide/message-api.html#guide-message-passing-api

https://docs.dgl.ai/guide/message-heterograph.html

Apply Message Passing On Part Of The Graph

https://docs.dgl.ai/guide/message-part.html

message function,Reduce function

https://docs.dgl.ai/api/python/dgl.function.html

https://docs.dgl.ai/api/python/nn.html

实现了常见的图卷积和图池化

以graphconv为例，不仅给出了接口和代码还有论文介绍

PyG， DGL对比

https://www.zhihu.com/question/399802947